[HIVE-14152] datanucleus.autoStartMechanismMode should set to 'Ignored' to allow rolling downgrade

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.1.1, 2.2.0
• Component/s: Metastore
• Labels: TODOC2.1.1 TODOC2.2
• Target Version/s: 1.3.0, 2.1.1, 2.2.0
• Assignee: Thejas Nair
• Reporter: Daniel Dai
• Created: 01/Jul/16 23:48
• Updated: 21/Oct/17 00:58
• Resolved: 13/Jul/16 17:03

Description

We see the following issue when downgrading metastore:
1. Run some query using new tables
2. Downgrade metastore
3. Restart metastore will complain the new table does not exist

In particular, constaints tables does not exist in branch-1. If we run Hive 2 and create a constraint, then downgrade metastore to Hive 1, datanucleus will complain:

javax.jdo.JDOFatalUserException: Error starting up DataNucleus : a class "org.apache.hadoop.hive.metastore.model.MConstraint" was listed as being persisted previously in this datastore, yet the class wasnt found. Perhaps it is used by a different DataNucleus-enabled application in this datastore, or you have changed your class names.
	at org.datanucleus.api.jdo.NucleusJDOHelper.getJDOExceptionForNucleusException(NucleusJDOHelper.java:528)
	at org.datanucleus.api.jdo.JDOPersistenceManagerFactory.freezeConfiguration(JDOPersistenceManagerFactory.java:788)
	at org.datanucleus.api.jdo.JDOPersistenceManagerFactory.createPersistenceManagerFactory(JDOPersistenceManagerFactory.java:333)
	at org.datanucleus.api.jdo.JDOPersistenceManagerFactory.getPersistenceManagerFactory(JDOPersistenceManagerFactory.java:202)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:498)
	at javax.jdo.JDOHelper$16.run(JDOHelper.java:1965)
	at java.security.AccessController.doPrivileged(Native Method)
	at javax.jdo.JDOHelper.invoke(JDOHelper.java:1960)
	at javax.jdo.JDOHelper.invokeGetPersistenceManagerFactoryOnImplementation(JDOHelper.java:1166)
	at javax.jdo.JDOHelper.getPersistenceManagerFactory(JDOHelper.java:808)
	at javax.jdo.JDOHelper.getPersistenceManagerFactory(JDOHelper.java:701)
	at org.apache.hadoop.hive.metastore.ObjectStore.getPMF(ObjectStore.java:377)
	at org.apache.hadoop.hive.metastore.ObjectStore.getPersistenceManager(ObjectStore.java:406)
	at org.apache.hadoop.hive.metastore.ObjectStore.initialize(ObjectStore.java:299)
	at org.apache.hadoop.hive.metastore.ObjectStore.setConf(ObjectStore.java:266)
	at org.apache.hadoop.util.ReflectionUtils.setConf(ReflectionUtils.java:76)
	at org.apache.hadoop.util.ReflectionUtils.newInstance(ReflectionUtils.java:136)
	at org.apache.hadoop.hive.metastore.RawStoreProxy.<init>(RawStoreProxy.java:60)
	at org.apache.hadoop.hive.metastore.RawStoreProxy.getProxy(RawStoreProxy.java:69)
	at org.apache.hadoop.hive.metastore.HiveMetaStore$HMSHandler.newRawStore(HiveMetaStore.java:650)
	at org.apache.hadoop.hive.metastore.HiveMetaStore$HMSHandler.getMS(HiveMetaStore.java:628)
	at org.apache.hadoop.hive.metastore.HiveMetaStore$HMSHandler.createDefaultDB(HiveMetaStore.java:677)
	at org.apache.hadoop.hive.metastore.HiveMetaStore$HMSHandler.init(HiveMetaStore.java:484)
	at org.apache.hadoop.hive.metastore.RetryingHMSHandler.<init>(RetryingHMSHandler.java:77)
	at org.apache.hadoop.hive.metastore.RetryingHMSHandler.getProxy(RetryingHMSHandler.java:83)
	at org.apache.hadoop.hive.metastore.HiveMetaStore.newRetryingHMSHandler(HiveMetaStore.java:5905)
	at org.apache.hadoop.hive.metastore.HiveMetaStore.newRetryingHMSHandler(HiveMetaStore.java:5900)
	at org.apache.hadoop.hive.metastore.HiveMetaStore.startMetaStore(HiveMetaStore.java:6159)
	at org.apache.hadoop.hive.metastore.HiveMetaStore.main(HiveMetaStore.java:6084)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:498)
	at org.apache.hadoop.util.RunJar.run(RunJar.java:221)
	at org.apache.hadoop.util.RunJar.main(RunJar.java:136)

Apparently datanucleus cache some trace about the new table and retry to reinstantiate later. This breaks downgrading we shall disable this behavior.

We need to set "datanucleus.autoStartMechanismMode" to "Ignored" to disable the check since it becomes a norm in downgrading case.

Issue Links

• breaks: HIVE-17872 Ignoring schema autostart doesn't work (HIVE-14152 used the wrong setting)
• is related to: HIVE-6113 Upgrade DataNucleus [was: Unable to instantiate org.apache.hadoop.hive.metastore.HiveMetaStoreClient]
• is related to: HIVE-14080 hive.metastore.schema.verification should check for schema compatiblity
• is related to: HIVE-1841 datanucleus.fixedDatastore should be true in hive-default.xml